Skip to content

Benchmark

MAST (Medical AI Superintelligence Test)

Scores are percentages. Higher is better.

Models measured
8
Scores recorded
8
Sources
1
IQ weight
0.80

Score distribution

58%
Mean
58.9%
Median
2%
Std dev
56.6%
10th percentile
60.1%
90th percentile
53.7%
Min
53.7%60.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.80 for chat models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-5.6 Solclinical-result-106· Clinical Benchmarks data docs60.2%
2Kimi K3clinical-result-107· Clinical Benchmarks data docs60.1%
3Gemini 3.6 Flashclinical-result-108· Clinical Benchmarks data docs59.3%
4Gemini 3.1 Pro Previewclinical-result-109· Clinical Benchmarks data docs58.9%
5Qwen3.5 397B A17Bclinical-result-110· Clinical Benchmarks data docs57.9%
6Claude Opus 5clinical-result-111· Clinical Benchmarks data docs57.1%
7Claude Sonnet 5clinical-result-112· Clinical Benchmarks data docs56.6%
8Grok 4.3clinical-result-113· Clinical Benchmarks data docs53.7%

Data credits: Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence)